How much soil do young children ingest: an epidemiologic study.

Sixty-four children aged 1-4 years were evaluated for the extent to which they ingest soil. The study followed the soil tracer methodology of S. Binder, D. Sokal, and D. Maughan (1986, Arch. Environ. Health, 41, 341-345). However, the present study included a number of modifications from the Binder et al. study. The principal new features were (1) increasing the tracer elements from three to eight; (2) using a mass-balance approach so that the contribution of food and medicine ingestion would be considered; (3) extending the period of observation from 3 days to 8 days; and (4) validating the methodology by having adult volunteers ingest known amounts of soil in a mass-balance validation study. The principal findings reveal the following. (1) The adult study confirmed the validity of the tracer methodology to estimate soil ingestion. (2) Of the eight tracers employed in the adult study, only Al, Si, and Y provided sufficient recovery data that was directly acceptably stable and reliable. (3) If food ingestion determinations were taken into consideration, the median estimates of soil ingestion from the eight tracers ranged from a low of 9 mg/day (Y) to a high of 96 mg/day (V); the median values of Al, Si, and Y, the three most reliable tracers, ranged from 9 mg/day to 40 mg/day. (4) One child had soil ingestion values ranging from 5 to 8 g/day, depending on the tracer. (5) If food ingestion had not been considered, the estimates of soil ingestion would have increased about two- to sixfold, depending on the tracer with Ti and Y being most affected by food intake. (6) Since soil and dust samples did not significantly differ in their levels of tracer elements, no reliable differentiation between the contribution of ingestion of dust and soil could be made. (7) These findings are generally consistent with the previously reported findings of Binder et al. (1986) and P. Clausing, B. Brunekreff, and J.H. van Wijnen (1987, Int. Arch. Occup. Med., 59, 73) if these latter studies are corrected for ingestion of tracers in food and medicine. The findings also account for the apparent discrepancy between the estimates from Al and Si and estimates based on Ti in previous studies. Thus the elevated estimates of soil ingestion by Ti were substantially reduced when food ingestion is considered.